Bug 64181
| Summary: | REGRESSION (r90602): Gmail doesn't load | ||
|---|---|---|---|
| Product: | WebKit | Reporter: | Ryosuke Niwa <rniwa> |
| Component: | JavaScriptCore | Assignee: | Nobody <webkit-unassigned> |
| Status: | RESOLVED FIXED | ||
| Severity: | Critical | CC: | abarth, ap, barraclough, eae, fpizlo, maccinema, mjs, ojan, oliver, sam |
| Priority: | P1 | ||
| Version: | 528+ (Nightly build) | ||
| Hardware: | Unspecified | ||
| OS: | Unspecified | ||
| Bug Depends on: | 64279 | ||
| Bug Blocks: | |||
Ryosuke Niwa
On ToT WebKit, Gmail no longer loads properly. Even if it loaded, it doesn't handle clicking and other events properly.
| Attachments | ||
|---|---|---|
| Add attachment proposed patch, testcase, etc. |
Ryosuke Niwa
I can't tell whether this is a regression in loader (XHR?), JSC, or event handling. The fact page doesn't finish loading seems to suggest some regression in loader but when it loads, I can't interact with the page properly (CSS styles seem to work properly) so I'm also suspecting JSC regression. However, gmail relies heavily on XHR so it's possible that user interactions are prevented due to some bugs in XHR.
Ryosuke Niwa
Okay, this regression still exists in r90643.
Adam Barth
Can we get a tighter regression range?
Ryosuke Niwa
(In reply to comment #3)
> Can we get a tighter regression range?
I'm building r90594 because I suspect that http://trac.webkit.org/changeset/90595/ is the cause.
Ryosuke Niwa
It was not r90595.
Alexey Proskuryakov
Testing with our archive builds says that it's r90602 indeed.
Ryosuke Niwa
(In reply to comment #6)
> Testing with our archive builds says that it's r90602 indeed.
I have confirmed this on my local build as well.
Ryosuke Niwa
+eae since he knows how gmail works.
Gavin Barraclough
*** Bug 64176 has been marked as a duplicate of this bug. ***
Gavin Barraclough
This should be fixed in r90673 - reverting the changes that look like they are to blame.